dlmypr's profile - activity

When I run tod_detecting I get the following error;

recognizer: /opt/ros/diamondback/stacks/geometry/eigen/include/Eigen/src/Core/DenseStorage.h:71: Eigen::internal::plain_array<t, size,="" matrixorarrayoptions,="" 16="">::plain_array() [with T = float, int Size = 4, int MatrixOrArrayOptions = 0]: Assertion `(reinterpret_cast<size_t>(array) & 0xf) == 0 && "this assertion is explained here: " "http://eigen.tuxfamily.org/dox/UnalignedArrayAssert.html" " * READ THIS WEB PAGE !!! *"' failed.

When I get a backtrace I see the problem is coming from the point_cloud.h inside the pcl package.

#0  0xb7fe1424 in __kernel_vsyscall ()
#1  0xb524c941 in raise () from /lib/libc.so.6
#2  0xb524fe42 in abort () from /lib/libc.so.6
#3  0xb52458e8 in __assert_fail () from /lib/libc.so.6
#4  0xb7f6703d in plain_array (this=0x8148de8) at /opt/ros/diamondback/stacks/geometry/eigen/include/Eigen/src/Core/DenseStorage.h:71
#5  DenseStorage (this=0x8148de8) at /opt/ros/diamondback/stacks/geometry/eigen/include/Eigen/src/Core/DenseStorage.h:104
#6  PlainObjectBase (this=0x8148de8) at /opt/ros/diamondback/stacks/geometry/eigen/include/Eigen/src/Core/PlainObjectBase.h:354
#7  Matrix (this=0x8148de8) at /opt/ros/diamondback/stacks/geometry/eigen/include/Eigen/src/Core/Matrix.h:272
#8  Quaternion (this=0x8148de8) at /opt/ros/diamondback/stacks/geometry/eigen/include/Eigen/src/Geometry/Quaternion.h:252
#9  Identity (this=0x8148de8) at /opt/ros/diamondback/stacks/geometry/eigen/include/Eigen/src/Geometry/Quaternion.h:109
#10 PointCloud (this=0x8148de8) at /opt/ros/diamondback/stacks/perception_pcl/pcl/include/pcl/point_cloud.h:84
#11 tod::TexturedObject::TexturedObject (this=0x8148de8) at /host/Desktop/Thesis/ros/repository/object_recognition/tod/src/core/TexturedObject.cpp:18
#12 0xb5551ca0 in tod::Loader::readTexturedObjects(std::vector<cv::Ptr<tod::TexturedObject>, std::allocator<cv::Ptr<tod::TexturedObject> > >&) ()
   from /host/ros/repository/object_recognition/tod_detecting/lib/libtod_detecting.so
#13 0x0805f7cf in ?? ()
#14 0xb5238ce7 in __libc_start_main () from /lib/libc.so.6
#15 0x0805d001 in ?? ()

I visited the web page http://eigen.tuxfamily.org/dox/Unalig... but i didn't understand what is causing this error.

Does anyone know how to cope with it?

Best regards..

Hi everyone,

Does anyone know what rvec and tvec corresponds to? I read the corresponding opencv pages but it just indicates that rvec is the rotation and tvec is the translation vector. But with respect to what?

Best regards..

References:

SolvePnp

FindExtrinsiccameraparams2

Hi everyone,

Does anyone know how to record a tod_training bag file. I have some unsuccessful attempts. I don't know why my bag files are not valid. I changed all the topics names of my kinect output topics as /camera_info /image (initial name ../rgb/image_color) /image_mono /points2 as my main bag file i record these topics to main.bag In main.tf.bag I have my tf message. I edit config.txt having the name main

But it doesn't take the image and point cloud instances as it should be.

How can I record a tod_training bag with kinect? What are the bottlenecks?

Best regards..

When we record a bag file for example X.bag and it contains the topic data /a_topic

Is there any way to rename the topics name inside the X.bag /a_topic to /b_topic?

Best regards..

Hi all,

Does anyone know a working package for retrieving pose (3D position) of an object from an image or bag file? I used Objects of Daily Use Finder (ODUfinder) and tod_training + tod_detecting pair. They work quite good but I can't retrieve the pose of the objects in the images or the input from a camera.

The two packages can recognize the objects existing in database, tod_detect also gives an estimation of the POSE but JUST VISUALLY. So somehow it computes the pose as it can give it visually but I don't know how I can get it numerically.

Thank you for your help,

Best regards..

Hi all,

I am using tod_training and tod_detecting packages for object detection and recognition. They work well but the output is just an image of estimated pose and the name of the recognized object.

My question is how can I find this estimated pose? Not just by seeing on the image.